Cost-effective optimal design of a pump-and-treat system for remediating groundwater contaminant at an industrial complex

Abstract

Trichloroethylene (TCE) and other solvents are found as groundwater contaminants in industrial complexes. Pump-andtreat (PAT) systems are commonly selected for remediation, but have the disadvantages of long operation time and high cost. To reduce the remediation time and cost of PAT systems, a cost-effective optimal design was obtained with a simulation–optimization model. MODFLOW and MT3DMS were used for simulating groundwater flow and contaminant transport. A genetic algorithm was used for optimization, and a parallel computing technique was used for faster optimization. Cost-effective optimal design was achieved using two objective functions, which calculated the total remediation cost by means of installation, operational, and maintenance costs. Based on the restrictions in the study area, two remediation scenarios were examined. Scenario 1 involved the removal of TCE, and Scenario 2 involved the containment of TCE within the compliance line. The total remediation costs of scenario 2 were 77.4~90.4% of scenario 1. The remediation time decreased from 9.87 years in scenario 1 to 8.13 years in scenario 2. Sensitivity analysis determined the effects of installation, operational, and maintenance costs on the total remediation cost, showing that optimal total cost and remediation time were most affected by operational cost. A parallel computing technique using 40 networked slave processes increased optimization speed by 16.9 times. However, the efficiency of the parallel computing technique decreased as networking time increased.